Computer Implemented Inventions - A Further Fallacy at the EPO
In the 2013 decision T 1670/07-Shopping with mobile device/NOKIA, Technical Board of Appeal 3.5.01 described a “gallery” of fallacies: frequently advanced arguments that appear reasonable but are fundamentally flawed. In decisions T 1755/10-Software structure/TRILOGY of 6 November 2014 and T 1834/10-Image selection/EBAY of 25 February 2015 the Board has added another, which they refer to as “the software implementation fallacy”. This argument seeks to suggest that “modified software causes a modified behaviour of the computer and should for that very reason (eo ipso) be considered as a technical implementation”. The Board counters that if the modified software implements a non-technical idea, its modification does not get over the hurdle of providing a further technical effect beyond that of the mere running of the program.
As is well known, non-technical inventions are not patentable in Europe. The Boards of Appeal of the EPO have developed a well worn approach to the examination of inventions that involve non-technical subject matter. Under the EPO’s approach, non-technical features in a claim are ignored and novelty and inventive step is judged on whatever technical features are left. For example, if a claim includes a feature such as “processor means configured to calculate a price according to a business rule”, the EPO ignores the details of the calculation performed and reads the feature as a generic processor.
The EPO’s approach has been justified on several grounds: that inventive step is judged by a technically skilled person; that non-technical features are part of the background to the invention; and that an invention must have a technical effect beyond the mere operation of the computer.
In Shopping with mobile device/NOKIA, Board of Appeal 3.5.01, which hears most appeals relating to computer-implemented inventions, summarised three arguments that applicants frequently advance unsuccessfully:
- "technical leakage fallacy", in which it is argued that the intrinsic technical nature of the implementation leaks back into the intrinsically non-technical nature of the problem
- "broken technical chain fallacy" in which a chain of steps from providing information to its use in a technical process is broken by the intervention of a user: the possible final technical effect brought about by the action of a user cannot be used to establish an overall technical effect because it is conditional on the mental activities of the user
- "non-technical prejudice fallacy" in which non-technical aspects are invoked as a reason for not modifying the prior art, whereas these features cannot in fact contribute to inventive step.
In Software structure/TRILOGY, a fourth fallacy is added; the argument that a new program provides a new technical effect merely by its execution. The application in suit concerned calculation of commission to be paid in relation to transactions and the applicants sought to argue that it provided a technical solution by virtue of the specific structure claimed. There was no attempt to argue that calculation of commissions was technical in nature. The claimed structure involved a data model including two objects "in the sense of object oriented programming" and a "commission engine". This structure was said to make it easy to reprogram the system if the rules for paying commission were changed, to allow parallel processing and to allow large numbers of transactions to be processed.
The Board noted that the overall goal of the invention ‘is a commercial goal; sales and marketing considerations ("commissions", "promotions", "reward", "credit", "compensation") cannot enter into the examination for an inventive step’.
Therefore an inventive step could lie only in the structure of the data model and engine, divorced from their functions. The Board held that ‘in the absence of any other potential "further" technical effect, the mere use of a specific software solution does not amount to a technical implementation’. The applicant's argument that the structure itself was technical and could be applied to other business methods was rejected as fallacious as it would make the exclusion from patentability of computer programs per se toothless.
The invention in Image selection/EBAY related to the selection of a predetermined number of images from a database of images to populate a dynamic webpage. A "utility program" was used to select the images, randomly in a specific embodiment. The appellant's arguments centred on assertions of the advantages of dynamic selection of images and that the claimed invention was simple and efficient. However the Board held that "the desire to present a selection of images, to limit the selection to a predetermined number of images and to present the selected images in some preferred layout is a non-technical aim."
Thus a technical invention could only lie in the details of the implementation and these were sparse in the claims and even the description. Referring to Software structure/TRILOGY, the Board stated that the "combination of two types of non- inventions (presentation of information, computer program) is not enough to define a technical contribution" and rejected all the appellant's requests.
These cases well illustrate the approach of the EPO to computer-implemented inventions and the great difficulty an applicant faces if it is asserted that the overall aim of an invention is non-technical. If a truly new and inventive data or program structure has general applicability it ought to be patentable, but if initially claimed in the context of a specific, non-technical implementation it is very hard to overcome. These cases also illustrate the difficulty of arguing that an implementation is inventive if details of the implementation are sparse. An appropriate level of detail needs to be included at the drafting stage and boilerplate clauses indicating that the invention can be carried out by any suitably programmed computing device tend to hinder rather than help.
24 April 2015